ATP ↔Pi exchange and membrane phosphorylation in sarcoplasmic

Jan 8, 1975 - E., Piglet, V. P., Gerhart, J. C., and Schach- man, . K.(1972), Biochemistry 11, 315. Perez-Polo, J. R., Bamburg, J. R., De Jong, W. W. ...
0 downloads 0 Views 677KB Size
E X C H A N G E BY SILVER

A C T I V A T I O N O F ATP-Pi

Biol. 12, 8 8 . Moore, Jr., J. B., Mobley, W. C., and Shooter, E. M. (1974), Biochemistry 13, 833. Nelbach, M. E., Pigiet, V. P., Gerhart, J. C., and Schachman, H. K. (1972), Biochemistry 1 1 , 315. Perez-Polo, J. R., Bamburg, J. R., De Jong, W. W. W., Straus, D., Baker, M., and Shooter, E. M. (1972b), in Nerve Growth Factor and Its Antiserum, E. Zaimis, Ed., London, Athlone Press, University of London, p 19. Perez-Polo, J. R., De Jong, W. W. W. Straus, D., and Shooter, E. M. (1972a), Adv. Exp. Med. Biol. 32, 91. Schenkein, I. (1972), in Handbook of Neurochemistry, Vol.

5B, Lajtha, A., Ed., New York, N.Y., Plenum Press, p 503. Schlesinger, M. J. ( 1 9 6 9 , J. Biol. Chem. 240, 4293. Scott, D. A. (1 934), Biochem. J. 28, 1592. Smith, A. P. (1969), Doctoral Thesis, Stanford University. Smith, A. P., Varon, S., and Shooter, E. M. (1968), Biochemistry 7 , 3259. Varon, S., Nomura, J., and Shooter, E. M. (1967), Biochemistry 6. 2202. Varon, S., Nomura, J., and Shooter, E. M. (1968). Biochemistry 7, 1296. Zeppezauer, M. (1971), Methods Enzymol. 22, 253.

-

ATP Pi Exchange and Membrane Phosphorylation in Sarcoplasmic Reticulum Vesicles: Activation by Silver in the Absence of a Ca2+ Concentration Gradient? Leopoldo de Meis* and Martha M. Sorensonl

-

ABSTRACT: The activation of ATP Pi exchange, normally associated with a Ca2+ concentration gradient in sarcoplasmic reticulum vesicles, can be obtained in ‘‘leaky’’ vesicles in 4-10 m M CaC12. In the micromolar range, Ag+ activates the ATP Pi exchange two- to fourfold. Similar concentrations of Ag+ promote a parallel inhibition of Ca2+activated ATP hydrolysis and Ca2+ uptake in intact ves-

icles. Maximal inhibition of these activities by Ag+ leaves the Mg2+-dependent ATPa’se unaffected. No net synthesis of ATP was demonstrated in leaky vesicles, The effects of Ag+ depend on the protein concentration and persist after removal of Ag+ from the medium. Membrane phosphorylation from Pi or from ATP is respectively activated or inhibited by Ag+ in reciprocal fashion.

S a r c o p l a s m i c reticulum vesicles (SRV)’ isolated from skeletal muscle actively take up Ca2+.from the medium in the presence of ATP and Mg2+. Hydrolysis of ATP and the translocation of Ca2+ into the vesicles involve a transfer of the y-phosphate of ATP to a membrane protein (E), forming an acylphosphoprotein (E-P). Accordingly, the following reaction sequence has been proposed (Makinose, 1969; Hasselbach, 1972).

chemical synthesis of ATP (Barlogie et al., 1971; Makinose, 197 1, 1972, 1973; Makinose and Hasselbach, 197 1 ; Hasselbach et al., 1972; Panet and Selinger, 1972; Deamer and Baskin, 1972; Masuda and de Meis, 1973, 1974). The following data support this conclusion. (a) Net Synthesis of A T P . When SRV previously loaded with calcium phosphate are incubated in a medium containing EGTA,’ Mg2+, ADP, and Pi, a fast release of Ca2+ coupled with ATP synthesis from ADP and Pi is observed (Makinose, 1972, 1973; Makinose and Hasselbach, 1971). (b) A T P Pi Exchange. When intact SRV are incubated in a medium containing ATP, Mg2+, [32P]Pi,and Ca2+, calcium phosphate is accumulated by the vesicles and a Ca2+ concentration gradient is built up until a steady state is reached in which a slow Ca2+ efflux is balanced by an ATP-driven influx. When this condition is reached, a steady rate of exchange between Pi and the y-phosphate of ATP is observed (Makinose, 1971; Racker, 1972; de Meis and Carvalho, 1974). This exchange implies that the two reactions shown above are operating simultaneously forward (ATP hydrolysis) and backward (ATP synthesis from ADP and Pi). (c) If the SRV are made “leaky” by various means, the Ca2+ concentration gradient is abolished and both the net synthesis of ATP and ATP Pi exchange reaction are arrested (Makinose, 1971; de Meis and Carvalho, 1974). When SRV are loaded with calcium phosphate, the formation of a gradient coincides with the establishment of three conditions: a high Ca2+ concentration (in the millimo-

-

E

+

c 2+

ATP-E-P E - P ---L E

+

+

ADP

Pi

Recently, it has been demonstrated that the entire process of Ca2+ transport can be reversed and that the enzymatic system of the SRV membrane is able to use the energy derived from a Ca2+ concentration gradient for the

From the Instituto de Biodsica, Universidade Federal do Ria de Janeiro, Centro de Ciencias MCdicas, Cidade Universitiria, Ilha do Fundgo, Ria de Janeiro, GB, Brazil. Received January 8, 1975. This investigation was supported in part by the Conselho Nacional de Pesquisas, B r a d (T.C. 17.1 17), by the Conselho de Ensino para Graduados da U.F.R.J., and by the Banco Nacional de Desenvolvimento Econjmico (FUNTEC 241). Recipient of a fellowship from the Conselho Nacional de Pesquisas, Brazil. Present address: Laboratory of Neurophysiology, Department of Neurology, College of Physicians and Surgeons of Columbia University, New York, N.Y. 10032. Abbreviations used are: SRV, sarcoplasmic reticulum vesicles; EGTA, ethylene glycol bis(P-aminoethyl ether)-N,N’-tetraacetic acid.

*



-

-

BIOCHEMISTRY.

VOL.

1 4 ,N O . 1 2 , 1 9 7 5

2739

DE M E I S AND SORENSON

lar range) inside the vesicles, a Ca2+ concentration in the micromolar range at the external surface of the vesicles, and a decrease in the rate of ATP hydrolysis due to the low Ca2+ concentration remaining in the assay medium (Hasselbach, 1972). In a previous study (de Meis and Carvalho, 1974), it was shown that solubilized or leaky SRV were still Pi exchange in the presence of a able to catalyze an ATP high Ca2+ concentration (1-8 mM). However, in these conditions the exchange rate was three to five times lower than that measured with intact SRV with a Ca2+ concentration gradient. In this paper it is shown that Ag+ inhibits Ca2+ uptake and the Ca2+-dependent ATPase of leaky SRV and at optimal Ca2+ concentrations activates two- to fourfold the rate Pi exchange. Some of the data were presented in of ATP less detail at a symposium on biological membranes in Rio de Janeiro in June, 1974.

.-c

E

io.20

-

-

W

ka 0

E

-

+' N

-

s t: 0.0-

i \

0 0.4 1

4 IO 20 50

/ U M &No3

Materials and Methods FIGURE

pL\q-3

1: Inhibition of Ca2+ uptake and ATPase activity of SRV by

Sarcoplasmic reticulum vesicles were prepared from rabAg+. For the Ca2+ uptake (left), the assay medium contained intact bit skeletal muscle as previously described (de Meis and vesicles and 30 m M Tris-maleate buffer (pH 7.0). 20 mM MgC12, 5 mM ATP, 8 m M Pi, and 0.1 mM 45CaC12. For the total ATPase activHasselbach, 1971). [ T - ~ ~ P I A Twas P prepared (de Meis, ity (right), leaky SRV, nonradioactive Ca, and [ Y - ~ ~ P ] A T were P used. 1972) using [32P]Piobtained from the Brazilian Institute of The reaction was started by the addition of 0.05 (V)or 0.20 (A)mg of Atomic Energy and purified by means of a column of SRV protein/ml. After 4 min incubation at 30' the reaction was Dowex A G l - X I 0 (de Meis and Carvalho, 1974). stopped either by the removal of SRV with Millipore filters or by the addition of trichloroacetic acid. The values shown in the figure repreATP Pi exchange was assayed by measuring the forsent the average f the standard error of seven (v)or four (A)experimation of [y-32P]ATP by the method of Avron (1960), ments. slightiy modified (de Meis and Carvalho, 1974). As a control, after extraction by the Avron method, the [ Y - ~ ~ P ] A T P Martonosi, 1970; de Meis and Carvalho, 1974), the leaky formed in the different experimental conditions described in vesicles no longer accumulated Ca2+, although the ATPase the Results section was identified enzymatically (Panet and activity is increased. Ca2+ uptake was measured by the use Selinger, 1972) using hexokinase (EC 2.7.1 . I ) and glucoseof 45Ca and Millipore filters as previously described (de 6-phosphate dehydrogenase (EC 1.1. I .4.9.). Meis, 1969, 1971). ATPase activity was assayed by measuring the release of Pretreatment with AgNO3. I n most experiments, the ef[32P]Pi from [ Y - ~ ~ P ] A TThe P . [32P]P,was extracted from fect of AgNO3 was assessed by adding SRV to a complete the assay medium as phosphomolybdate with isobutyl alcoassay medium containing the AgNO3. I n some experiments, hol-benzene (de Meis and Carvalho, 1974). Two different ATPase activities can be distinguished in SRV. The Mg2+- the SRV were tested after preincubation with AgN03 at concentrations specified in the results: SRV (0.35-0.5 mg dependent ATPase requires only Mg2+ for its activation of protein/ml) were incubated for 3-5 min at room temperand is tested in the presence of EGTA to remove contaminating Ca2+. The Ca2+-activated ATPase requires Ca2+ ature in 5 m M Tris-maleate buffer (pH 7.0) containing AgNO3. The pellet formed after 30 min of centrifugation at and Mg2+ for full activity and is calculated by subtracting 20,OOOg was resuspended in 5 m M Tris-maleate buffer (pH the Mg2+-dependent activity from the total activity mea7.0) before being added to the assay testing medium. Essensured in the presence of Mg2+ and Ca2+ (see Table I). Only tially the same results were obtained if the pellet was the Ca*+-activated ATPase is intimately associated with washed once with 40 ml of buffer before being resuspended SRV Ca2+ transport (Hasselbach, 1972), and the rate of for use. If the pretreatment with AgNO3 was performed in this ATPase depends on the Ca2+ concentration in the a more complete assay medium, the SRV behaved quite difassay medium. In experiments in which the total ATPase ferently (see text). activity is measured as a function of the incubation time, Membrane phosphorylation from either [ T - ~ ~ P I A TorP the removal of Ca2+ from the assay medium by SRV results [32P]Pi was assayed as previously described and corrected in a progressive decrease of the rate of the Ca2+-activated for nonspecific binding (de Meis and Masuda, 1974). component. Therefore, in order to obtain linear rates, all ATPase activities were measured using leaky SRV (Fiehn Results and Hasselbach, 1969). Inhibition of Ca2+ Transport by AgNO3. Micromolar Preparation of Leaky SRV. A suspension of SRV at 9 to concentrations of AgN03 pramote a parallel inhibition of 12 mg of protein/ml in 1 to 2 m M EGTA was adjusted to the Ca2+ uptake and the ATPase activity of SRV (Figure pH 9.2 by the addition of 1 M Tris. After 20 min at room 1). This inhibition is observed when either oxalate or phostemperature, the pH was readjusted to 7.0 by the addition phate is present in the assay medium to increase the Ca2+ of Tris-maleate buffer (pH 6 . 0 ) , and the suspension was concentrating ability of the SRV (Hasselbach, 1972; de used immediately. This treatment increases the Ca*+ perMeis et al., 1974). For both the Ca2+ uptake and the ATmeability of the SRV membranes, allowing any Ca2+ which Pase activity, the concentration of AgN03 required for 50% might have accumulated inside the vesicles to flow out. As a inhibition increases with the SRV protein concentration control, Ca2+ uptake and Ca2+-dependent ATPase activity (Figure 1). The data of Table I show that in the range of were measured in several preparations before and after this AgNO3 concentrations used, only the Ca2+-activated ATtreatment. In agreement with earlier reports (Duggan and

-

2740

BIOCHEMISTRY, VOL.

14, NO. 1 2 , 1 9 7 5

A C T I V A T I O N O F ATP-Pi

E X C H A N G E BY S I L V E R

Table I: Effect of Ag+ on SRV ATPase Activity and ATP

-

Pi Exchange.a

-

ATP PjExchange, / m o l of ATP * Pi/mg of Protein. 5 min

ATPase Act.. Ltmol of Pdmg of Protein, 5 min ~

Additions to Assay Medium or SRV Treatment (A) None (B) AgNO, added to assay medium (5 X M) (C) SRV preincubated in 5 X lo-’ MAgNO,

~~

Mgz+ -Dependent EGTA

Low Caz+

High CaZf

Low CaZf

0.98 0.84

0.05 f 0.02

10.7 0.87

f i

0.4 0.04

2.25 r 0.08 0.92 r 0.08

9.71 + 0.40 0.03 0.02

1.09 + 0.13

1.11

?

0.12

1.28 r 0.11

0.02

f

Total

CaZ+-Activated

+_

* 0.02

High C a z + 1.47 0.08

f +_

0.18 0.08

Low C a z + 0 0.01

+_

0.01

0.08 + 0.08

High Ca’+ 0.10 * 0.01 0.42 t 0.04 0

a A and B are paired experiments on four preparations of leaky SRV at 30” mM Tris-maleate buffer (pH 7.0), 5 mM ATP, 0.5 mM ADP, 6 rnM Pi, 20 mM MgCl,, and 5 mM EGTA or 0.1 mM CaC1, or 4.0 mM CaC1,. The Caz+-activated ATPase was calculated for each experiment as described under Materials and Methods. Tne experiments in C were performed o n 11 preparations of SRV pretreated with AgNO, and buffer as described under Materials and Methods, and tested a t 30” in an assay medium without AgNO, but containing 1 mM ATP, 1 mM MgCl,, 20 mM KC1, 20 mM NH,Cl, and 30 mMTris-maleate buffer (pH 7.0). For the MgZfdependent activity the EGTA concentration was 2 mM. Low CaZtwas 30 @ ’I (0.5 mM EGTA and 0.47 mM CaCl,) and high Ca2+was 5 mMCaC1,. Reactions were started by the addition of SRV (0.2 or 0.3 mg of protein/ml) and stopped after 5 min with trichloroacetic acid. ATPase activity and ATP -Pi exchange were determined as described under Materials and Methods, using parallel assays containing [r-”P] ATP or [”PI Pi, respectively. Values represent the average the standard error. +_

z

2-

I-

03

c E

-

0

P

zW

P

!5

fb B

E

02-

p

v)

dI

-

l -

n

0‘

v)

\

0 A

s

-

3

0-

0

2

4

6

8

IO

MINUTES

2: Time course of activation of ATP Pi exchange by Ag+. All assays contained 30 mM Tris-maleate buffer (pH 7.0), 5 mM ATP, 0.5 m M ADP, 6 mM [32P]Pi, 20 mM MgC12, and 4 m M CaC12, at 30’. The reactions were started by the addition of leaky SRV (0.3 mg of protein/ml) and aliquots were mixed with trichloroacetic acid at the times indicated. Formation of [y3*P]ATP was analyzed as described under Materials and Methods: ).( control (no AgN03); (0) 20 bM AgNO3 added before SRV; (A,o)20 fiM AgNO3 added 3 or 6 min after SRV, as shown by arrows. FIGURE

++

Pase is inhibited, while the M g 2 + - d e p e n d e n t ATPase is not modified.

-

FIGURE 3: Effects of Ag+ on exchange and hydrolysis at different SRV concentrations. The Pi concentration was 8 m M and AgN03 was present as indicated. Other components of the assay medium and experimental conditions were as described for the control medium in Figure 2. The figure shows a typical experiment. With 0.1 mg of leaky SRV protein/ml, the concentration of AgNO3 required for maximum activation of the exchange reaction varied from 4 to 20 X M in five different SRV preparations tested: (left) ATP Pi exchange; (right) total ATPase activity; (A) leaky SRV, 0.1 mg of protein/ml; (V)0.3 mg of protein/ml.

-

Pi Exchange Reaction by In c o n t r a s t t o t h e inhibitory effect shown in F i g u r e 1 , AgN03 activates ATP Pi exchange in leaky vesicles in the presence of a high Ca2+ concentration. In the experim e n t of F i g u r e 2, t h e addition o f 20 p M A g N 0 3 a u g m e n t ed t h e rate of e x c h a n g e to similar levels w h e t h e r it w a s present in t h e m e d i u m before the SRV or added several m i n u t e s after the reaction had begun. The degree of activation a t optimal concentrations of AgN03 w a s similar to t h a t seen w h e n comparing leaky vesicles w i t h i n t a c t vesicles having a Ca2+ concentration g r a d i e n t ( T a b l e 11). The rate of ATP Pi e x c h a n g e as well as t h e d e g r e e of its activation

concentration ( F i g u r e 3 , left). In t h e e x p e r i m e n t s of F i g u r e 1, a parallel inhibition of the ATPase activity a n d Ca2+ upt a k e was shown. A similar parallelism w a s not f o u n d in leaky vesicles in the presence of a high Ca2+ concentration for the effects on ATPase activity a n d ATP Pi e x c h a n g e . With 0.1 mg of protein/ml, for e x a m p l e ( F i g u r e 3 ) , t h e ATP Pi exchange reaction w a s fully activated at 4 p M AgN03, w h e r e a s the ATPase activity w a s not modified. When the AgNO3 w a s raised to 10 a n d 20 wM, the r a t e of exchange remained high but the total ATPase activity w a s inhibited to the level of the M g 2 + - d e p e n d e n t ATPase. With

by AgN03 varied considerably a m o n g preparations, even at the same protein concentration (cf. T a b l e s I a n d 11). H o w ever, t h e addition of AgNO3 at its o p t i m a l concentration t o the assay m e d i u m never failed to activate ATP Pi exc h a n g e at least twofold. P o t a s s i u m n i t r a t e in the range of 1 - 100 W Mh a d n o effect o n e x c h a n g e . T h e concentration of A g + required for m a x i m a l activa-

AgNO3 activated hydrolysis and higher concentrations in-

Activation of the A T P

-

AgN03.

-

-

-

-

higher concentrations of S R V , low concentrations hibited it.

of

-

Ca2+ Concentration Dependence. In a previous report ( d e Meis and Carvalho, 1974), no ATP Pi e x c h a n g e w a s m e a s u r e d in leaky SRV in the presence of 0.1 m M CaZ+, w h e r e a s ATP hydrolysis w a s fully activated. R a i s i n g t h e Ca2+ concentration of t h e m e d i u m from 0.1 to 8 m M proBIOCHEMISTRY, VOL.

14, N O . 1 2 , 1 9 7 5

2741

D E MEIS A N D SORENSON

5

0.5

1

0.4

-

IO

z

E

10

'

E

103-

E '

tw

B

a

6. 02-

E" 1

01-

z>

5 v)

$

'

c

f

4 .

.

0-

-

-

mM CoCf 2

4: Ca2+dependence of ATP P, exchange and ATPase activity in the presence of AgN03. For ATP P, exchange (left), the assay medium composition was 30 mM Tris-maleate buffer (pH 6.8). 5 m M ATP, 0.5 mM ADP, 6 rnM [32P]PI,20 mM MgC12, and different CaC12 concentrations as shown in the figure. For zero Ca2+, CaC12 was omitted and 5 mM EGTA was added. For the total ATPase activity (right), nonradioactive Pi and [y-32P]ATPwere used. The reaction was started by the addition of leaky SRV (0.2 or 0.3 mg of protein/ml) and stopped after 5-min incubation at 30' by the addition of trichloroacetic acid. The values shown represent the average the standard error of four experiments: control, without AgNO3 (V);with 5 X MAgNO3 (A). FIGURE

*

port and Pi liberation (Yamamoto and Tonomura, 1967; Makinose, 1969; de Meis and de Mello, 1973). In a previous report, little phosphorylation from [32P]Piwas found in the presence of Ca2+ and ATP (de Meis and Masuda, pmol of ATP Pijmg 1974). Apparently, the phosphorylation sites are preferenSRV Treatment of Protein, 5 min tially phosphorylated by ATP (de Meis, 1972; de Meis and (A) Leaky 0.064 t 0.01 3 (5) de Mello, 1973). Under similar conditions (Figure 5 ) , the (B) Leaky + AgNO, 0.202 t 0.033 ( 5 ) membrane phosphorylation from [ Y - ~ ~ P I A T is Pinhibited (C) Intact. with gradient 0.200 + 0.034 (4) ~by AgNO3. Concomitantly, membrane phosphorylation a The assay medium contained 30 mM Tris-maleatebuffer (pH from [32P]Piis enhanced. These effects are seen both a t low 7.0), 5 mM ATP, 20 mM MgCl,, 6 mM ["PI Pi, 4 mM CaC1, (A and and a t high Ca2+concentrations, although only in the latter B), or 0.1 mkf CaCl, (C), and 0.2 mg of SRV protein/ml, a t 30". A and B contained 0.5 mM ADP. For experiment C, ADP was omitted case does AgNO3 activate ATP Pi exchange (Figure 4). and the rate of ATP + + P iexchange was measured after 99% of the Preincubation of SRV with AgNO3. The aim of the folCa2+ of the assay medium had been removed by the SRV (de Meis lowing set of experiments was to ascertain whether the efand Carvalho, 1974). For B, each preparation was tested in a range fect of AgNO3 is related to the formation in the assay mediof AgNO, concentrations as in Figure 3, and the maximal rate of exchange in each case was used to alculate the values shown. The um of different substrate species, such as a silver-nucleovalues represent the average * the standard error of the number of tide complex, or whether it is consistent with binding of experiments indicated in parentheses. Ag+ to the S R V membrane. The dependence of the effects of AgNO3 on the SRV protein concentration (Figures 1 and 3 ) lends support to the second possibility. gressively inhibited ATPase activity, a t the same time actiI n the first set of experiments, S R V were preincubated as vating ATP PI exchange. In a similar experiment (Figure described under Materials and Methods in a buffer solution 4 and Table I), the addition of 50 p M AgN03 fully inhibitto containing different AgN03 concentrations ( ed the Ca2+-activated ATPase a t all Ca2+ concentrations, M ) before testing the rates of exchange and hydrolysis in 4 while the ATP PI exchange reaction was activated twom M CaC12 under the conditions of the control experiments to fourfold a t Ca2+ concentrations above 1 mM. Both in the to 4 X M of Figure 4. In vesicles pretreated with presence and in the absence of AgN03, the Ca2+ concenAgN03, neither the exchange reaction nor the Ca2+-actitration required for 50% of the maximal activation of the vated ATPase activity was modified. With pretreatment in exchange reaction ranged from 1.7 to 2.4 mM. higher A g N 0 3 concentrations, the Mg2+-dependent ATP, Concentration Dependence. The rate of A T P PI exPase remained unimpaired but both of the other activities change varies with the P, concentration of the assay mediwere progressively inhibited, being fully inhibited a t 5 X um (de Meis and Carvalho, 1974). The addition of 50 p M M AgNO3, These data reinforce those of Table I Ag+ activated exchange in ten preparations under condishowing that the Ca2+-activated ATPase is more sensitive tions similar to those of Figure 2 using from 1 to 8 m M P,. to A g N 0 3 than is the Mg2+-dependent ATPase. Incubating This is the maximum of PI that can be used with 4 m M SRV with buffer and AgN03 alone did not produce the acCaC12 without causing precipitation of calcium phosphate. Pi exchange reaction seen when tivation of the A T P The average activation was threefold a t 1 m M P, and twoA g N 0 3 was added to the complete assay medium (Table I ) . fold at 8 m M P,. In a second set of experiments, leaky SRV were preincuMembrane Phosphorylation f r o m ATP and P , . In the bated with AgNO3 in a medium that was optimal for the process of ATP hydrolysis by the Ca2+-activated ATPase, exchange reaction (Figure 4) except that it lacked Mg2+. the y-phosphate of A T P is covalently bound to a membrane After centrifugation and resuspension of the pellet in the protein. This phosphoprotein represents an intermediate same medium but without AgN03, the activity of the prepproduct in the sequence of reactions leading to Ca2+ transTable 11: Comparison of Activation of ATP +Pi Exchange by AgNO, with That by a CaZf Concentration Gradient.a +

-

-

-

-

-

2742

BIOCHEMISTRY, VOL.

14,

NO.

1 2 , 1975

A C T I V A T I O N O F ATP-Pi

I

EXCHANGE BY SILVER

4mMcoa;

0.1mM CaC12



Y

0

1

410

I

W

4000

O L -4

/YM A0 NO3

o

5: Membrane phosphorylation by [32P]Pi and [y-”P]ATP. The assay medium composition was 30 m M Tris-maleate buffer (pH 7.0), 5 m M ATP, 0.5 m M ADP, 6 m M Pi, 20 mM MgCI2, and 0.1 (left) or 4.0 m M (right) CaC12. The reaction was started by the addition of leaky S R V (0.33 mg of protein/ml) and stopped after 20- or 40-sec incubation at 30° by the addition of trichloroacetic acid to a final concentration of 20% (w/v). The degree of membrane phosphorylation from either compound was essentially the same for either of these times, indicating that the steady state was reached within the first 20 sec: ( 0 ) [ Y - ~ ~ P J A and T P nonradioactive Pi; (0) [32P]Pi and nonradioactive ATP. The values shown in the figure represent the average & the standard error of four or seven experiments, respectively.

f

2

/y M & N O 3



4 0

*om

-

-

0.3

iE

L

4 IO

I 04

.€

2

0.41

F I G U R E 7 : Effect of AgNO3 on S R V preincubated with AgN03. Preincubation of leaky SRV with AgNO,, assay medium, and experimental conditions were as described for Figure 6 , differing only in the addition of AgNO3 to the assay medium in the concentrations shown. The values represent the average & the standard error of four experiments. The concentrations of AgNO, in the preincubation were: (0) M. zero (control); (A) lo-’ M;and (0)2 X

-e

0 - 1 ’ o m

at

/y4w33

FIGURE

r J

02-

,urn c m 4

0

w

Rmcuklm J

FIGURE 6 : Preincubation of S R V with AgNO3. The assay medium contained 30 mM Tris-maleate buffer (pH 7.0), 5 mM ATP, 0.5 mM ADP, 6 mM Pi, 4 mM CaC12, and 20 mM MgC12. Leaky S R V at 0.5 mg of protein/ml were preincubated for 3 min at room temperature in the same medium but without MgCl2 as described in the text. ATP P, exchange and total ATPase activity were tested at a final concentration of about 0.5 mg of protein/ml. After 5-min incubation at 30° the reaction was arrested by the addition of trichloroacetic acid. Other conditions are described under Materials and Methods. The values shown in the figure represent the average f the standard error of four experiments: (V)ATP PI exchange; the medium contained [32P]P, and nonradioactive ATP; (A)total ATPase activity; the medium contained [ Y - ~ ~ P I A and T P nonradioactive P,.

-

-

aration was tested with Mg2+ added t o the assay medium. Figure 6 shows that the exchange reaction was activated and the ATPase inhibited by the pretreatment with A g N 0 3 (cf. Figure 3). Figure 7 shows that when S R V were treated with the optimal concentration of AgNO3 in the preincubation, subsequent addition of AgNO3 to the assay medium

-

8: Effect of CuSO4 on the ATP Pi exchange. The assay medium contained 30 mM Tris-maleate buffer (pH 7.0), 5 m M ATP, 2 m M ADP, 8 m M [32P]Pi, 20 m M MgC12, 4 m M CaC12, and the specified concentrations of CuSO4. The reaction was started by the addition of leaky S R V (0.3 mg of protein/ml) and stopped after 4-min incubation at 3 7 O by the addition of trichloroacetic acid. Essentially the same results were obtained in three different S R V preparations tested. FIGURE

-

did not activate further the rate of A T P Pi exchange. Similarly, SRV which were treated with an excessive concentration of A g N 0 3 in the preincubation were no longer activated by the addition of AgN03 to the assay medium. These data and those of Figures 1 and 3 lead to the conclusion that the effect of AgNO3 is consistent with its binding to a membrane component, and that one or more ligands, except Mg2+, involved in the exchange reaction (ATP, ADP, Ca2+, and Pi) somehow protect the enzyme from a nonspecific inhibitory effect of AgN03. Specificity. Copper sulfate is also able to activate the ATP Pi exchange, although somewhat less effectively than A g N 0 3 (Figure 8). K N 0 3 , zinc acetate, FeC13,

-

BIOCHEMISTRY,VOL.

14,

NO.

12, 1975

2743

DE MEIS AND SORENSON

-

FeS04, PtC16, coC12, PdC12, and CdC12 in the concentration range to M had no effect on the ATP PI exchange reaction. Discussion

-

In this paper it is shown that the binding of Ag+ to leaky SRV membranes activates the rate of ATP PI exchange to a level similar to that measured in vesicles having a transmembrane Ca2+ concentration gradient. This indicates that the gradient is not necessary as an energy source for the reaction (Makinose, 1971; Makinose and Hasselbach, 1971; de Meis and Carvalho, 1974), but probably favors one or more of the intermediate steps involved in the overall exchange reaction. Exchange of phosphate between the medium and ATP implies formation of a phosphoprotein from P,, but it has not been possible to demonstrate this correlation in intact vesicles in the presence of ATP and Ca2+ (Kanazawa and Boyer, 1973; de Meis and Masuda, 1974). Figure 6 shows that this can be done in the presence of Ag+ and provides additional information which can be related to the exchange reaction in untreated vesicles. (a) The phosphorylation from PI was equally activated by Ag+ at a low and at a high Ca2+ concentration. Therefore, the role of a high Ca2+ concentration in promoting the exchange reaction (Figure 4) is not related to the phosphorylation step, but to the transfer of this phosphate to ADP. This implies that formation of E-P is followed by formation of Ca:E-P. We are presently attempting to distinguish these two forms of the phosphoprotein by other means. (b) The activation by Ag+ of the membrane phosphorylation from PI coincides with an inhibition of the membrane phosphorylation from ATP. Evidence has been presented that ATP and PI are substrates of a common site in the membrane and that the binding of Ca2+ to its outer surface dictates a preference for ATP in the phosphorylation reaction (Kanazawa and Boyer, 1973; Masuda and de Meis, 1973; de Meis and Masuda, 1974). The binding of Ag+ apparently can override this control mechanism and imitate the effect of low external Ca2+ concentrations in intact vesicles by favoring the phosphorylation from PI, decreasing phosphorylation from ATP, and consequently inhibiting the Ca2+-dependent ATPase activity. With the information presented here and in previous papers (Kanazawa and Boyer, 1973; Masuda and de Meis, 1973; de Meis and Masuda, 1974; de Meis and Carvalho, 1974; Kanazawa, 1975) the following hypothesis is proposed. The binding of Ca2+ to the internal surface of the membrane allows the transfer of the phosphate from the phosphoenzyme to ADP, while the binding of Ca2+ to the PI outer surface of the SRV modulates the rate of ATP exchange by regulating the number of sites phosphorylated from ATP or P,. The Ca2+ binding site on the outer surface of membrane would be saturated in micromolar Ca2+ concentrations (Makinose, 1969; Hasselbach, 1972; de Meis and de Mello, 1973), while the binding site at the inner surface would only be saturated in the millimolar range (de Meis and Carvalho, 1974). Finally, the Ca2+-dependent ATPase of SRV is similar to the (Na+ K+)-ATPase provided that Ca2+ and N a + are assigned corresponding roles for the two enzymes. Both enzymes are phosphorylated from ATP or P,, both catalyze ADP ATP, H O H PI, and ATP PI exchanges, and both enzymes have a similar primary structure in the neighborhood of the active site of phosphorylation (Bastide et al.,

-

+

-

2744

-

-

BIOCHEMISTRY,

VOL.

14, NO.

1 2 , 1975

1973; Dahms and Boyer, 1973; de Meis and Masuda, 1974; Hasselbach, 1972; Kanazawa and Boyer, 1973; Masuda and de Meis, 1973; Post et al., 1975: Taniguchi and Post, 1975). Acknowledgments We are grateful to Dr. Robert L. Post for a stimulating discussion of the manuscript. The excellent technical assistance of Mr. Isaltino R. Soares and Mr. Valdecir Antunes Suzano is acknowledged with thanks. Ref eren ces Avron, M. (1960), Biochim. Biophys. Acta 40, 257. Barlogie, E., Hasselbach, W., and Makinose, M. (1971), FEBS Lett. 12, 267. Bastide, F., Meissner, G., Fleischer, S., and Post, R. L. ( 1 973), J. Biol. Chem. 248, 8485. Dahms, A. S., and Boyer, P. D. (1973), J. Biol. Chem. 248, 3 155. Deamer, D. W., and Baskin, R. J. (1972), Arch. Biochem. Biophys. 153, 47. de Meis, L. (1969), J . Biol. Chem. 244, 3733. de Meis, L. (1971), J. Biol. Chem. 246, 4764. de Meis, L. (1972), Biochemistry 11, 2460. de Meis, L. and Carvalho, M. G. C. (1974), Biochemistry 13, 5032. de Meis, L., and de Mello, M. C. F. (1 973), J . Biol. Chem. 248, 3691. de Meis, L., and Hasselbach, W. (1971), J . Biol. Chem. 246, 4759. de Meis, L., Hasselbach, W., and Machado, R. D. (1974), J . Cell Biol. 62, 505. de Meis, L., and Masuda, H . (1974), Biochemistry 13, 2057. Duggan, P. F., and Martonosi, A. (1970), J . Gen. Physiol. 56, 147. Fiehn, W., and Hasselbach, W. (1969), Eur. J . Biochem. 9, 574. Hasselbach, W. (1972), in Molecular Bioenergetics and Macromolecular Biochemistry, Weber, H . H., Ed., New York, N.Y., Springer-Verlag, p 149. Hasselbach, W., Makinose, M., and Migala, A. (1972) FEBS Lett. 20, 3 1 1. Kanazawa, T. (1975), J . Biol. Chem. 250, 113. Kanazawa, T., and Boyer, P. D. (1973), J . Biol. Chem. 248, 3163. Makinose, M. (1969), Eur. J . Biochem. 10, 74. Makinose, M. (1971), FEBS Lett. 12, 269. Makinose, M. (1972), FEBS Lett. 25, 1 13. Makinose, M. (1973), FEBS Lett. 37, 140. Makinose, M., and Hasselbach, W . (1971), FEBS Lett. 12, 271. Masuda, H., and de Meis, L. (1973), Biochemistry 12, 458 1. Masuda, H., and de Meis, L. (1974), Biochim. Biophys. Acta 332, 313. Panet, R., and Selinger, Z . (1972), Biochim. Biophys. Acta 255, 34. Post, R. L., Toda, G., and Rogers, F. N . (1975), J . Biol. Chem. 250, 691. Racker, E. (1972), J . Biol. Chem. 247, 8198. Taniguchi, K., and Post, R. L. (1975), J . Biol. Chem. (in press). Yamamoto, T., and Tonomura, Y. (1967), J . Biochem. (Tokyo)62, 558.